My copending U.S. patent application Ser. No. 10/124,689; filed Apr. 12, 2002, entitled MULTI-FUNCTIONAL MICRO ELECTROMECHANICAL DEVICES AND METHOD OF BULK MANUFACTURING SAME discloses and claims a method of bulk manufacturing SiC sensors, including pressure sensors and accelerometers. The disclosure of my copending application is incorporated herein by reference.
I am a named inventor of U.S. Pat. No. 5,637,905 to Carr et al. and it discloses a high temperature pressure and displacement microsensor made from a Si substrate. A first coil structure is positioned within a recess in the Si substrate. A pressure diaphragm is glass bonded about its periphery to the rim of the recess in the semiconductor substrate. A second coil structure is positioned on the underside of the pressure diaphragm and is electrically isolated from the first coil structure. The coils are inductively coupled together and provide an output indicative of changes in the coupling between the coils.
My U.S. Pat. No. 6,248,646 discloses a process for making an array of SiC wafers on standard larger industry sized wafers. This patent discusses the operating conditions for SiC and SiC-On-Insulator technology and cites the need for sensors made from SiC.
U.S. Pat. No. 5,973,590 to Kurtz et al. discloses a hermetically sealed semiconductor sensor bonded to first and second glass wafers.
U.S. Pat. No. 6,319,757 B1 to Parsons et al. discloses a silicon-carbide wafer bonded to an underlying ceramic substrate. The '757 patent states at col. 6, lns. 7-13 that the silicon carbide semiconductor substrate is a SiC die and the underlying substrate is polycrystalline aluminum nitride. Parsons et al. further states that the expansion coefficients of aluminum nitride and silicon carbide are nearly identical in the aforementioned structure and that of the borosilicate (BSG) glass encapsulant is close enough to both silicon carbide and aluminum nitride so as to avoid separation or cracking over a wide temperature range. Parsons et al. also teaches a flip chip application in FIG. 3 thereof encapsulated in glass.
U.S. Pat. No. 5,891,751 to Kurtz et. al describes curing a glass frit bonding the cover to the transducer. Upon curing, the glass frit becomes the peripheral glass layer 106. During the curing process, gasses are created which escape through an aperture designed for the purpose of the escaping gasses. According to the '751 patent, the aperture prevents the glass frit from bubbling and out gassing during the curing process which would prevent a hermetic seal along the periphery of the structure. The aperture accordingly must be within the inner periphery of the glass. See, the '751 patent at col. 7, Ins. 5-17. In FIG. 1 herein reference numeral 100 is used to indicate the prior art drawing FIG. 7 of the '751 patent. Referring to FIG. 1 herein, Kurtz et al. identify transducer 101 which includes silicon (Si) diaphragm 103 and dielectric 102 (silicon dioxide). Glass 106 bonds silicon (Si) cover member 105 to the transducer (sensor). Aperture 108 permits out gassing during curing so as to not ruin the glass bond 106 and the seal it makes. Piezoresistors 109 reside on the dielectric 102. Glass bottom cover 104 includes aperture 107. In FIG. 2 herein, reference numeral 200 identifies the prior art drawing FIG. 8 of the '751 patent illustrating electrostatic bonding of glass sheet 201 to the top cover 105 to seal aperture 108 under vacuum conditions. Contact pads 202 are exposed in the prior art.
FIG. 3 herein is a duplicate of FIG. 4 of U.S. Pat. No. 6,058,782 to Kurtz et al. Reference numeral 300 signifies one of the hermetically sealed ultra high temperature silicon carbide pressure transducers of the '782 patent which has been cut or diced. See, col. 3, lns. 52-54. Referring to FIG. 3, silicon carbide first substrate 302 having unnumbered piezoresistors thereon bonded to silicon carbide second substrate 303 by electrostatic bonding or by glass frit bonding. See, col. 6, lns. 9-23 of the '782 patent. If bonded by a glass frit the same is not illustrated in the '782 patent as no space is shown between substrates 302 and 303 in FIG. 3. Further no provision or illustration is made for the escape aperture which must reside in second substrate 303 which might be referred to herein as the cover. Sensor 302 and cover 303, bonded together are illustrated as being in engagement with header 304 which carries a glass insulator (unnumbered) bonded to the top cover. Referring again to FIG. 3, leads 301 are illustrated in engagement with a platinum glass frit 305 electrically communicating with contact pads 306.
Referring to FIG. 3, the second substrate 303 is mounted atop glass which is unnumbered in FIG. 3. Since the glass expands at a rate of thermal expansion which is different than the substrate 303, stress is applied to the second substrate which may cause the separation of the pins 301 from the contact pads 306 of the first substrate. Stress may also be applied to the piezoresistors on the first substrate inducing measurement error.
The use of electrostatic bonding method makes very weak bond strength between the SiC sensors and the SiC cover. This may lead to debonding during thermal cycling thereby rendering the device useless. Application of glass frits as the adhesion material between the SiC cover and the SiC sensor makes necessary the creation of an aperture as an escape path for out gassing during glass bonding. Since the aperture will have to be sealed later in order to maintain the desired hermetic reference cavity, it increases the risk of the sealant sipping into the reference cavity.
There is growing demand for improved efficient management of energy consumption in jet engines and automobiles. Global reduction of undesirable emissions of hydrocarbons and other combustion by-products such as oxides of nitrogen and carbon monoxide are being sought assiduously. Semiconductor based sensors and electronics targeted for insertion in high temperature, extreme vibration, and corrosive media must satisfy a set of minimum reliability criteria before becoming acceptable for operational use. In addition, it is crucial to validate the Computational Fluid Dynamics codes generated for flow fields and turbulent conditions inside engines. Validation of these codes is necessary to render them trustworthy. Devices capable of functioning in these harsh environments need the appropriate package to sustain stable and reliable operation during the life of the device. Package reliability problems have largely contributed to prevent the application of these devices.
Typically these devices operate in environments of 300° C. and above. This is very challenging since conventional semiconductor electronic and sensing devices are limited to operating in temperatures less than 300° C. due to the limitations imposed by material properties and packaging. Silicon carbide-based electronics and sensors have been demonstrated to operate in temperatures up to 600° C. thereby offering promise of direct insertion into the high temperature environment. However, the lack of the device (sensor) packaging methodologies appropriate for this harsh environment has affected the operational reliability and survivability of these devices (sensors). Reliability problems at high temperature due to poor packaging has discouraged global application and large-scale commercialization of these devices. As such, the much anticipated introduction of SiC devices into high temperature environments has been delayed.